INDEX

Explanations

time-related expressions, such as durations and time indicators

oai_token-act-pair · gpt-3.5-turbo

temporal references related to durations and intervals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olis

-0.69

emale

-0.65

avid

-0.61

adium

-0.61

ghazi

-0.61

](

-0.60

[+

-0.60

liction

-0.59

icio

-0.59

 Shape

-0.57

POSITIVE LOGITS

 later

1.81

later

1.49

 Later

1.41

Later

1.38

 thereafter

1.28

 afterward

1.27

 afterwards

1.25

Ago

1.08

 subsequent

1.07

ago

1.03

Activations Density 0.090%

time-related expressions, such as durations and time indicators

temporal references related to durations and intervals

No Comments

No Known Activations